On 2026/10/05 13:04, Enzo Nicosia wrote: > Hi all, > > sorry but have to post this here as net/iamb is maintained by the ports > ML: > > net/iamb segfaults in the latest 8.0 snapshot. I have built it from the > latest port tree. The system is running the latest snapshot (updated on > October 3rd). I have not been able to pinpoint the issue though, as the > stack trace is not exactly "legible". The latest few entries are as > follows:
Steps to reproduce would be helpful. > ..... > #41 0x000009ceb4593fff in > _RINvNtNtCs5OGfc95HXmw_3std3sys9backtrace28___rust_begin_short_backtraceNCNvMs4_NtNtNtCsg65Z0yXabRi_5tokio7runtime8blocking4poolNtB1f_7Spawner12spawn_thread0uEB1l_ > () from /usr/local/bin/iamb backtraces start with the most recently called function and work backwards - the entries near the start are needed (though preferably the complete trace) > #42 0x000009ceb45a1b47 in > _RNSNvYNCINvNtNtCs5OGfc95HXmw_3std6thread9lifecycle15spawn_uncheckedNCNvMs4_NtNtNtCsg65Z0yXabRi_5tokio7runtime8blocking4poolNtB1b_7Spawner12spawn_thread0uEs_0INtNtNtCs9rH4Ric1Emj_4core3ops8function6FnOnceuE9call_once6vtableB1h_ > () from /usr/local/bin/iamb > #43 0x000009ceb48a324b in > _RNvNvMs0_NtNtNtCs5OGfc95HXmw_3std3sys6thread4unixNtB7_6Thread3new12thread_start > () from /usr/local/bin/iamb > #44 0x000009d171bbec52 in _rthread_start (v=Unhandled dwarf expression > opcode 0xf3) at /usr/src/lib/librthread/rthread.c:99 >From this 'unhandled dward expression' I suspect you are using gdb (the ancient version from base which doesn't work too well with more recent compilers) rather than egdb (from the gdb package). > #45 0x000009d10635be4a in __tfork_thread () at > /usr/src/lib/libc/arch/amd64/sys/tfork_thread.S:87 > #46 0x000009d10635be4a in __tfork_thread () at > /usr/src/lib/libc/arch/amd64/sys/tfork_thread.S:87 > Previous frame identical to this frame (corrupt stack?) > > I gather it is an issue with tfork_thread but have no clue why this > stopped working. I must admit there was a jump of about three weeks > since the previous sysupgrade -s here, so I am afraid I cannot narrow > this down further at this point. > > Willing to help with testing if needed. > > HND > > Enzo Nicosia > > -- >
